Dr. Stephen Henson [Wed, 24 Mar 2010 23:42:30 +0000 (23:42 +0000)]
initialise buf if wrong_info not used
Dr. Stephen Henson [Wed, 24 Mar 2010 23:16:35 +0000 (23:16 +0000)]
PR: 1731 and maybe 2197
Clear error queue in a few places in SSL code where errors are expected
so they don't stay in the queue.
Dr. Stephen Henson [Wed, 24 Mar 2010 13:16:55 +0000 (13:16 +0000)]
prepare for release
Dr. Stephen Henson [Wed, 24 Mar 2010 13:16:42 +0000 (13:16 +0000)]
Submitted by: Bodo Moeller and Adam Langley (Google).
Fix for "Record of death" vulnerability CVE-2010-0740.
Andy Polyakov [Mon, 22 Mar 2010 22:44:48 +0000 (22:44 +0000)]
rand_win.c: fix logical bug in readscreen [from HEAD].
Andy Polyakov [Mon, 22 Mar 2010 22:40:18 +0000 (22:40 +0000)]
bss_file.c: fix MSC 6.0 warning [from HEAD].
Andy Polyakov [Mon, 22 Mar 2010 20:58:43 +0000 (20:58 +0000)]
ppc.pl: assembler Y chokes on apostrophes in comment.
Andy Polyakov [Mon, 15 Mar 2010 22:30:09 +0000 (22:30 +0000)]
e_capi.c: fix typo [from HEAD].
Andy Polyakov [Mon, 15 Mar 2010 22:27:32 +0000 (22:27 +0000)]
Fix UPLINK typo [from HEAD].
Dr. Stephen Henson [Mon, 15 Mar 2010 13:12:00 +0000 (13:12 +0000)]
workaround for missing definition in some headers
Dr. Stephen Henson [Fri, 12 Mar 2010 12:48:56 +0000 (12:48 +0000)]
PR: 2192
Submitted By: Jaroslav Imrich <jaroslav.imrich@disig.sk>
The prompt_info and wrong_info parameters can be empty strings which
can produce confusing prompts. Treat empty string same as NULL.
Dr. Stephen Henson [Fri, 12 Mar 2010 12:07:16 +0000 (12:07 +0000)]
missing goto meant signature was never printed out
Dr. Stephen Henson [Wed, 10 Mar 2010 13:48:35 +0000 (13:48 +0000)]
don't leave bogus errors in the queue
Dr. Stephen Henson [Tue, 9 Mar 2010 17:08:24 +0000 (17:08 +0000)]
PR: 2186
Submitted By: "Joel Rabinovitch" <Joel.Rabinovitch@tecsys.com>
Detect aix64-gcc
Dr. Stephen Henson [Sun, 7 Mar 2010 16:40:31 +0000 (16:40 +0000)]
The OID sanity check was incorrect. It should only disallow *leading* 0x80
values.
Dr. Stephen Henson [Sat, 6 Mar 2010 20:52:33 +0000 (20:52 +0000)]
don't add digest alias if signature algorithm is undefined
Dr. Stephen Henson [Fri, 5 Mar 2010 13:35:06 +0000 (13:35 +0000)]
Fix memory leak: free up ENGINE functional reference if digest is not
found in an ENGINE.
Dr. Stephen Henson [Wed, 3 Mar 2010 19:56:00 +0000 (19:56 +0000)]
PR: 2183
PR#1999 broke fork detection by assuming HAVE_FORK was set for all platforms.
Include original HAVE_FORK detection logic while allowing it to be
overridden on specific platforms with -DHAVE_FORK=1 or -DHAVE_FORK=0
Dr. Stephen Henson [Wed, 3 Mar 2010 15:34:11 +0000 (15:34 +0000)]
Submitted by: Tomas Hoger <thoger@redhat.com>
Fix for CVE-2010-0433 where some kerberos enabled versions of OpenSSL
could be crashed if the relevant tables were not present (e.g. chrooted).
Dr. Stephen Henson [Wed, 3 Mar 2010 15:30:05 +0000 (15:30 +0000)]
don't mix definitions and code
Andy Polyakov [Tue, 2 Mar 2010 16:26:13 +0000 (16:26 +0000)]
Fix s390x-specific HOST_l2c|c2l [from HEAD].
Submitted by: Andreas Krebbel
Dr. Stephen Henson [Mon, 1 Mar 2010 23:54:19 +0000 (23:54 +0000)]
PR: 2178
Submitted by: "Kennedy, Brendan" <brendan.kennedy@intel.com>
Handle error codes correctly: cryptodev returns 0 for success whereas OpenSSL
returns 1.
Dr. Stephen Henson [Mon, 1 Mar 2010 00:30:11 +0000 (00:30 +0000)]
load cryptodev if HAVE_CRYPTODEV is set too
Ben Laurie [Sun, 28 Feb 2010 13:37:15 +0000 (13:37 +0000)]
Fix warnings.
Dr. Stephen Henson [Fri, 26 Feb 2010 19:56:10 +0000 (19:56 +0000)]
quote HOSTCC in case it isn't defined
Dr. Stephen Henson [Fri, 26 Feb 2010 14:41:38 +0000 (14:41 +0000)]
Revert CFB block length change. Despite what SP800-38a says the input to
CFB mode does *not* have to be a multiple of the block length and several
other specifications (e.g. PKCS#11) do not require this.
Dr. Stephen Henson [Fri, 26 Feb 2010 14:34:24 +0000 (14:34 +0000)]
Change versions for 0.9.8n-dev
Dr. Stephen Henson [Thu, 25 Feb 2010 17:18:23 +0000 (17:18 +0000)]
Prepare for 0.9.8m release
Richard Levitte [Wed, 24 Feb 2010 01:20:04 +0000 (01:20 +0000)]
Since crypto-lib.com is built to be executed in the crypto/ directory,
there's no need to specify that directory in the include path.
Dr. Stephen Henson [Tue, 23 Feb 2010 14:09:32 +0000 (14:09 +0000)]
The meaning of the X509_V_ERR_UNABLE_TO_GET_ISSUER_CERT_LOCALLY and
X509_V_ERR_UNABLE_TO_GET_ISSUER_CERT error codes were reversed in
the verify application documentation.
Bodo Möller [Tue, 23 Feb 2010 10:36:41 +0000 (10:36 +0000)]
Always check bn_wexpend() return values for failure (CVE-2009-3245).
(The CHANGES entry covers the change from PR #2111 as well, submitted by
Martin Olsson.)
Submitted by: Neel Mehta
Richard Levitte [Tue, 23 Feb 2010 07:51:39 +0000 (07:51 +0000)]
Apply changes from the 1.0.0 branch.
Richard Levitte [Tue, 23 Feb 2010 07:50:54 +0000 (07:50 +0000)]
Include [.CRYPTO.<ARCH>] instead of just [.<ARCH>]
Richard Levitte [Mon, 22 Feb 2010 07:05:50 +0000 (07:05 +0000)]
In some environments, we need to defined sslroot locally.
Richard Levitte [Mon, 22 Feb 2010 07:05:24 +0000 (07:05 +0000)]
Add t1_reneg to the VMS build.
Hack the symbols with long names.
Bodo Möller [Fri, 19 Feb 2010 18:25:39 +0000 (18:25 +0000)]
Fix X509_STORE locking
Dr. Stephen Henson [Thu, 18 Feb 2010 12:42:03 +0000 (12:42 +0000)]
clarify documentation
Dr. Stephen Henson [Wed, 17 Feb 2010 19:43:08 +0000 (19:43 +0000)]
OR default SSL_OP_LEGACY_SERVER_CONNECT so existing options are preserved
Dr. Stephen Henson [Wed, 17 Feb 2010 18:37:47 +0000 (18:37 +0000)]
Allow renegotiation if SSL_OP_LEGACY_SERVER_CONNECT is set as well as
initial connection to unpatched servers. There are no additional security
concerns in doing this as clients don't see renegotiation during an
attack anyway.
Dr. Stephen Henson [Wed, 17 Feb 2010 14:32:01 +0000 (14:32 +0000)]
PR: 2100
Submitted by: James Baker <jbaker@tableausoftware.com> et al.
Workaround for slow Heap32Next on some versions of Windows.
Dr. Stephen Henson [Tue, 16 Feb 2010 14:19:42 +0000 (14:19 +0000)]
PR: 2171
Submitted by: Tomas Mraz <tmraz@redhat.com>
Since SSLv2 doesn't support renegotiation at all don't reject it if
legacy renegotiation isn't enabled.
Also can now use SSL2 compatible client hello because RFC5746 supports it.
Dr. Stephen Henson [Mon, 15 Feb 2010 19:40:45 +0000 (19:40 +0000)]
The "block length" for CFB mode was incorrectly coded as 1 all the time. It
should be the number of feedback bits expressed in bytes. For CFB1 mode set
this to 1 by rounding up to the nearest multiple of 8.
Dr. Stephen Henson [Mon, 15 Feb 2010 19:25:37 +0000 (19:25 +0000)]
Correct ECB mode EVP_CIPHER definition: IV length is 0
Dr. Stephen Henson [Mon, 15 Feb 2010 19:02:53 +0000 (19:02 +0000)]
PR: 2164
Submitted by: "Noszticzius, Istvan" <inoszticzius@rightnow.com>
Don't clear the output buffer: ciphers should correctly the same input
and output buffers.
Dr. Stephen Henson [Fri, 12 Feb 2010 22:02:07 +0000 (22:02 +0000)]
update references to new RI RFC
Dr. Stephen Henson [Tue, 9 Feb 2010 14:18:15 +0000 (14:18 +0000)]
Fix memory leak in ENGINE autoconfig code. Improve error logging.
Dr. Stephen Henson [Tue, 9 Feb 2010 14:13:30 +0000 (14:13 +0000)]
update year
Dr. Stephen Henson [Thu, 4 Feb 2010 01:10:24 +0000 (01:10 +0000)]
Only use bufferoverflowu.lib when needed
Dr. Stephen Henson [Tue, 2 Feb 2010 14:19:54 +0000 (14:19 +0000)]
tolerate broken CMS/PKCS7 implementations using signature OID instead of digest
Dr. Stephen Henson [Tue, 2 Feb 2010 14:03:07 +0000 (14:03 +0000)]
make no-rsa no-dsa compile again
Dr. Stephen Henson [Mon, 1 Feb 2010 16:48:40 +0000 (16:48 +0000)]
PR: 2160
Submitted by: Robin Seggelmann <seggelmann@fh-muenster.de>
Make session tickets work with DTLS.
Dr. Stephen Henson [Mon, 1 Feb 2010 12:44:21 +0000 (12:44 +0000)]
PR: 2159
Submitted by: Robin Seggelmann <seggelmann@fh-muenster.de>
Typo in PR#1949 bug, oops!
Dr. Stephen Henson [Thu, 28 Jan 2010 17:53:11 +0000 (17:53 +0000)]
In engine_table_select() don't clear out entire error queue: just clear
out any we added using ERR_set_mark() and ERR_pop_to_mark() otherwise
errors from other sources (e.g. SSL library) can be wiped.
Dr. Stephen Henson [Wed, 27 Jan 2010 18:53:59 +0000 (18:53 +0000)]
reword RI description
Dr. Stephen Henson [Wed, 27 Jan 2010 17:50:47 +0000 (17:50 +0000)]
update documentation to reflect new renegotiation options
Dr. Stephen Henson [Wed, 27 Jan 2010 16:06:36 +0000 (16:06 +0000)]
Some shells print out the directory name if CDPATH is set breaking the
pod2man test. Use ./util instead to avoid this.
Dr. Stephen Henson [Wed, 27 Jan 2010 14:04:51 +0000 (14:04 +0000)]
typo
Dr. Stephen Henson [Wed, 27 Jan 2010 14:03:26 +0000 (14:03 +0000)]
stop warnings in fips_test_suite application
Dr. Stephen Henson [Wed, 27 Jan 2010 13:32:31 +0000 (13:32 +0000)]
stop missing prototype warnings
Dr. Stephen Henson [Wed, 27 Jan 2010 13:21:34 +0000 (13:21 +0000)]
eliminate some warnings in fips build
Dr. Stephen Henson [Wed, 27 Jan 2010 00:51:24 +0000 (00:51 +0000)]
Bypass algorithm blocking with TLS MD5+SHA1 signature in FIPS mode by
calling underlying method directly.
Dr. Stephen Henson [Tue, 26 Jan 2010 19:40:36 +0000 (19:40 +0000)]
PR: 1949
Submitted by: steve@openssl.org
More robust fix and workaround for PR#1949. Don't try to work out if there
is any write pending data as this can be unreliable: always flush.
Dr. Stephen Henson [Tue, 26 Jan 2010 18:08:42 +0000 (18:08 +0000)]
PR: 2138
Submitted by: Kevin Regan <k.regan@f5.com>
Clear stat structure if -DPURIFY is set to avoid problems on some
platforms which include unitialised fields.
Dr. Stephen Henson [Tue, 26 Jan 2010 13:24:08 +0000 (13:24 +0000)]
Fix VC++ warning (change had already been made to other branches).
Dr. Stephen Henson [Tue, 26 Jan 2010 12:29:32 +0000 (12:29 +0000)]
Typo
Dr. Stephen Henson [Mon, 25 Jan 2010 16:08:52 +0000 (16:08 +0000)]
Update OID table too.
Dr. Stephen Henson [Mon, 25 Jan 2010 16:08:01 +0000 (16:08 +0000)]
PR: 2149
Submitted by: Douglas Stebila <douglas@stebila.ca>
Fix wap OIDs.
Dr. Stephen Henson [Sun, 24 Jan 2010 13:52:38 +0000 (13:52 +0000)]
oops revert test code from previous commit
Dr. Stephen Henson [Sun, 24 Jan 2010 13:50:57 +0000 (13:50 +0000)]
The fix for PR#1949 unfortunately broke cases where the BIO_CTRL_WPENDING
ctrl is incorrectly implemented (e.g. some versions of Apache). As a workaround
call both BIO_CTRL_INFO and BIO_CTRL_WPENDING if it returns zero. This should
both address the original bug and retain compatibility with the old behaviour.
Dr. Stephen Henson [Fri, 22 Jan 2010 18:49:43 +0000 (18:49 +0000)]
If legacy renegotiation is not permitted then send a fatal alert if a patched
server attempts to renegotiate with an unpatched client.
Dr. Stephen Henson [Wed, 20 Jan 2010 18:22:04 +0000 (18:22 +0000)]
change versions back to 0.9.8m-dev
Dr. Stephen Henson [Wed, 20 Jan 2010 17:26:02 +0000 (17:26 +0000)]
prepare for release
Dr. Stephen Henson [Wed, 20 Jan 2010 17:16:52 +0000 (17:16 +0000)]
update TABLE
Dr. Stephen Henson [Wed, 20 Jan 2010 16:35:30 +0000 (16:35 +0000)]
make update
Dr. Stephen Henson [Wed, 20 Jan 2010 14:04:29 +0000 (14:04 +0000)]
Support -L options in VC++ link.
Andy Polyakov [Tue, 19 Jan 2010 21:45:45 +0000 (21:45 +0000)]
rand_win.c: handel GetTickCount wrap-around [from HEAD].
Andy Polyakov [Tue, 19 Jan 2010 21:45:16 +0000 (21:45 +0000)]
x86_64-xlate.pl: refine sign extension logic when handling lea [from HEAD].
PR: 2094,2095
Dr. Stephen Henson [Tue, 19 Jan 2010 19:10:53 +0000 (19:10 +0000)]
revert patch
Dr. Stephen Henson [Tue, 19 Jan 2010 19:10:03 +0000 (19:10 +0000)]
PR: 2144
Submitted by: Robin Seggelmann <seggelmann@fh-muenster.de>
Better fix for PR#2144
Andy Polyakov [Sun, 17 Jan 2010 19:43:49 +0000 (19:43 +0000)]
Valgrind fix to aes-x86_64.pl in 0.9.8. For reference, newer aes-x86_64.pl
don't suffer from the problem after Win64 SEH support was added.
PR: 2075
Submitted by: Peter Klotz
Dr. Stephen Henson [Sat, 16 Jan 2010 19:45:46 +0000 (19:45 +0000)]
PR: 2144
Submitted by: steve@openssl.org
Fix DTLS connection so new_session is reset if we read second client hello:
new_session is used to detect renegotiation.
Dr. Stephen Henson [Sat, 16 Jan 2010 19:18:31 +0000 (19:18 +0000)]
PR: 2133
Submitted by: steve@openssl.org
Add missing DTLS state strings.
Dr. Stephen Henson [Sat, 16 Jan 2010 19:02:43 +0000 (19:02 +0000)]
Add strings for DTLS protocol versions
Dr. Stephen Henson [Thu, 14 Jan 2010 17:44:46 +0000 (17:44 +0000)]
PR: 1618
Submitted by: steve@openssl.org
Fix bug in 0.9.8-stable time handling in ca.c . NB: this only handles cases
where times are not being checked or printed properly. Issues relating to
time_t becoming negative or wrapping around are *NOT* addressed. OpenSSL
1.0.0 and later does fix these issues by using its own time routines.
Dr. Stephen Henson [Wed, 13 Jan 2010 19:08:45 +0000 (19:08 +0000)]
Fix version handling so it can cope with a major version >3.
Although it will be many years before TLS v2.0 or later appears old versions
of servers have a habit of hanging around for a considerable time so best
if we handle this properly now.
Dr. Stephen Henson [Wed, 13 Jan 2010 18:45:03 +0000 (18:45 +0000)]
Modify compression code so it avoids using ex_data free functions. This
stops applications that call CRYPTO_free_all_ex_data() prematurely leaking
memory.
Dr. Stephen Henson [Thu, 7 Jan 2010 19:09:32 +0000 (19:09 +0000)]
Simplify RI+SCSV logic:
1. Send SCSV is not renegotiating, never empty RI.
2. Send RI if renegotiating.
Andy Polyakov [Thu, 7 Jan 2010 11:22:25 +0000 (11:22 +0000)]
x86_64-xlate.pl: new gas requires sign extention in lea instruction
[from HEAD].
PR: 2094,2095
Andy Polyakov [Thu, 7 Jan 2010 11:04:49 +0000 (11:04 +0000)]
util/pl/VC-32.pl: bufferoverflowu.lib only when actually needed [from HEAD].
PR: 2086
Dr. Stephen Henson [Wed, 6 Jan 2010 18:02:07 +0000 (18:02 +0000)]
correct error codes
Dr. Stephen Henson [Wed, 6 Jan 2010 17:59:41 +0000 (17:59 +0000)]
Updates to conform with draft-ietf-tls-renegotiation-03.txt:
1. Add provisional SCSV value.
2. Don't send SCSV and RI at same time.
3. Fatal error is SCSV received when renegotiating.
Dr. Stephen Henson [Tue, 5 Jan 2010 17:50:12 +0000 (17:50 +0000)]
Typo
Dr. Stephen Henson [Tue, 5 Jan 2010 17:33:20 +0000 (17:33 +0000)]
PR: 2132
Submitted by: steve
Fix bundled pod2man.pl to handle alternative comment formats.
Dr. Stephen Henson [Sun, 27 Dec 2009 23:03:40 +0000 (23:03 +0000)]
Update RI to match latest spec.
MCSV is now called SCSV.
Don't send SCSV if renegotiating.
Also note if RI is empty in debug messages.
Dr. Stephen Henson [Fri, 25 Dec 2009 14:11:18 +0000 (14:11 +0000)]
Traditional Yuletide commit ;-)
Add Triple DES CFB1 and CFB8 to algorithm list and NID translation.
Bodo Möller [Tue, 22 Dec 2009 11:45:57 +0000 (11:45 +0000)]
Constify crypto/cast.
Bodo Möller [Tue, 22 Dec 2009 10:59:03 +0000 (10:59 +0000)]
Constify crypto/cast.
Dr. Stephen Henson [Thu, 17 Dec 2009 16:38:18 +0000 (16:38 +0000)]
Ooops, engage ENGINE initialisation code correctly in FIPS builds.
Dr. Stephen Henson [Thu, 17 Dec 2009 15:42:25 +0000 (15:42 +0000)]
Alert to use is now defined in spec: update code
Dr. Stephen Henson [Wed, 16 Dec 2009 20:34:20 +0000 (20:34 +0000)]
New option to enable/disable connection to unpatched servers